Infection by the hepatitis C virus (HCV) is responsible for most transfusion-associated cases of non-A, non-B hepatitis and also accounts for a significant proportion of community-acquired hepatitis cases worldwide. Relatively few HCV infected individuals experience acute hepatitis, but up to 85% appear to develop persistent infection that often leads to chronic hepatitis and liver cirrhosis, eventually predisposing them to hepatocellular carcinoma. At present, vaccines are unavailable and no broadly effective therapies exist for this viral disease. Consequently, much research has focused on the HCV replicative enzymes as targets for more effective therapies.
HCV contains an approximately 9.6 kb single-stranded positive sense RNA genome classified as its own genus in the Flaviviridae family of animal viruses, which also includes the flavivirus and pestivirus genera. Its genome consists of a conserved 5' nontranslated sequence that serves as an internal ribosome entry site, a single open reading frame that encodes a polyprotein of &gt;3000 amino acids, and a 3' nontranslated region. The 3' nontranslated region contains tracts of poly(U).sub.n and poly(UC).sub.n followed by a novel conserved 98 nucleotide sequence.
Proteolytic processing of the HCV polyprotein by virally-encoded proteases generates several nonstructural (NS) proteins with enzymatic activities essential for the replicative cycle of the virus [P. Neddermann et al., Biol. Chem., 378, pp. 469-476 (1997)]. NS2 encodes a presumed metalloprotease, NS5B is a RNA-dependent RNA polymerase, and NS3 is a bifunctional enzyme with a serine protease localized to the N-terminal 181 residues of the protein and a RNA helicase in the C-terminal 465 amino acids. The NS3 protease performs an intramolecular cleavage at the NS3/NS4A junction to form a tight noncovalent NS3-NS4A complex necessary for efficient processing of the remaining polyprotein [C. Failla et al., J. Virol., 69, pp. 1769-1777 (1995); R. Bartenschlager et al., J. Virol., 69, pp. 7519-7528 (1995); Y. Tanji et al., J. Virol., 69, pp. 1575-1581 (1995)]. To date, no evidence exists to suggest that the serine protease and helicase domains are separated by proteolytic processing of NS3 in vivo. This may reflect economical packaging of these enzymatic components, or could imply a functional interdependence between the two domains.
Numerous studies have demonstrated that the serine protease [J. L. Kim et al., Cell, 87, pp. 343-355 (1996); W. Markland et al., J. Gen. Virol., 78, pp. 39-43(1997).; C. Steinkuhler et al., J. Virol., 70, pp. 6694-6700 (1996)] and RNA helicase domains [J. A. Suzich et al., J. Virol., 67, pp. 6152-6158 (1993); C. L. Tai et al., J. Virol., 70, pp. 8477-8484 (1996); L. Jin et al., Arch. Biochem. Biophys., 323, pp. 47-53 (1995); and F. Preugschat et al., J. Biol. Chem., 271, pp. 24449-24457 (1996)] of NS3 can be expressed independently and isolated as catalytically active species. However, emerging evidence suggests that the NS3 protease and helicase domains may contact one another and modulate NS3 catalytic activities. Examples include apparent differences in pH optima of ATPase and RNA unwinding activities between a contiguous NS3 protein complexed with the NS4A cofactor [K. A Morgenstern et al., J. Virol., 71, pp. 3767-3775 (1997); Z. Hong et al., J. Virol., 70, pp. 4261-4268 (1996)] and an isolated NS3 helicase domain [C. L. Tai et al., J. Virol., (1996), supra; L. Jin et al., Arch. Biochem. Biophys., (1995), supra; F. Preugschat et al., J. Biol. Chem., (1996), supra; and Y. Gwack et al., Biochem. Biophys. Res. Commun., 225, pp. 654-659 (1996)]. Similarly, the ATPase activities of both proteins differ in their sensitivity to polynucleotide stimulation. Contiguous NS3 appears to have a lower apparent dissociation constant for poly(U) than does the helicase domain [J. A. Suzich et al., J. Virol., (1993), supra; F. Preugschat et al., J. Biol. Chem., (1996), supra; K. A. Morgenstern et al., J. Virol., (1997), supra; A. Kanai et al., FEBS Lett., 376, pp. 221-224 (1995)]. Aside from these differences, both proteins display nearly indistinguishable kinetic parameters for NTP hydrolysis when stimulated with saturating polynucleotide [J. A. Suzich et al., J. Virol., (1993), supra; K. A. Morgenstern et al., J. Virol., (1997), supra], both display 3'-5' directionality for translocation along a polynucleotide substrate, and the helicases of both proteins effectively unwind duplex RNA:RNA substrates [C. L. Tai et al., J. Virol., (1996), supra; Z. Hong et al., J. Virol., (1996), supra].
In addition to HCV, all flavi- and pestiviruses sequenced to date contain conserved helicase sequence motifs in their homologous NS3 proteins, suggesting that this enzyme plays an important role in the HCV replicative cycle [R. H. Miller et al., Proc. Natl. Acad. Sci. USA, 87, pp. 2057-2061 (1990)]. Consistent with this possibility, helicase encoding sequences have been identified in other viruses and helicases are suggested to catalyze the separation of double-stranded nucleic acid structures during transcription and genome replication [G. Kadare et al., J. Virol., 71, pp. 2583-2590 (1997)]. Previous studies with poliovirus, a positive-stranded RNA virus of the Picornaviridae family, show that mutation of conserved sequence motifs in the 2C elicase inhibits virus replication and proliferation [C. Mirzayan et al., Virology, 189, pp. 547-555 (1992)]. Similar mutational studies on the helicases encoded by herpes simplex virus type 1 and bovine papilloma virus also show that these enzymes are critical for virus replication [P. MacPherson et al., Virology, 204, pp. 403-408 (1994); R. Martinez et al., J. Virol., 66, pp. 6735-6746 (1992)]. Thus, the ability to inhibit helicase activity in HCV may provide an avenue for the therapeutic treatment of HCV infection.
Unfortunately, little is known about the details of how ATP binding and hydrolysis leads to DNA or RNA strand unwinding by the helicase. Two structures of helicases crystallized in the absence of polynucleotide, but, unfortunately, they have not yielded the critical information needed to extrapolate to an enzyme mechanism [N. Yao et al., Nat. Struct. Biol., 4, pp. 463-467 (1997); H. S. Subramanya et al., Nature, 384, pp. 379-383 (1996)].
Thus, there is a great need to solve the crystal structure of the helicase complexed with an oligonucleotide and, in particular, to delineate the oligonucleotide and nucleotide triphosphate (NTP) binding pockets of that enzyme. With this information, computer models of these binding sites can be created and potential inhibitors of HCV helicase can be rationally designed.